NVIDIA and MediaTek to build AI edge-to-cloud platforms

NVIDIA and MediaTek are collaborating on building next-generation AI computing platforms spanning cloud AI factories, local AI devices and automotive systems, with MediaTek adopting NVIDIA’s NVLink Fusion platform for custom AI accelerators in rack-scale AI factories.

MediaTek will work within NVIDIA’s NVLink Fusion ecosystem to help hyperscalers, cloud service providers and frontier model developers design custom XPUs that integrate with NVIDIA NVLink-connected, rack-scale AI factories.

The partners will continue co-developing multiple generations of NVIDIA RTX Spark and DGX Spark PC chips that pair NVIDIA GPUs with MediaTek system-on-chip (SoC) designs for consumer PCs, AI developer supercomputers and enterprise workstations.

They will keep advancing platforms for AI-powered, software-defined vehicles in the era of physical AI, building on existing integrations of MediaTek Dimensity Auto with NVIDIA technologies for intelligent cockpits alongside NVIDIA DRIVE AGX.

As part of the deal, NVIDIA has also invested US$3.5 billion in convertible bonds issued by MediaTek.

NVLink Fusion foundation

MediaTek will offer the NVLink Fusion platform as a pre-built, pre-qualified and system pre-validated foundation for multi-die XPU development to accelerate the path from silicon and advanced packaging to rack-scale systems. The platform bundles key technologies around a custom XPU, such as NVIDIA NVLink Fusion chiplet to connect XPUs to the NVIDIA NVLink scale-up fabric via photonics or electrical interconnects; NVIDIA NVLink-C2C for high-bandwidth, energy-efficient connectivity between XPUs, NVIDIA Rosa CPUs and other compatible processors; and NVIDIA NVHBM to integrate customised memory that increases bandwidth and energy efficiency while dedicating more silicon area to compute.

By leveraging this foundation, customers can focus on differentiated compute while relying on NVIDIA and MediaTek for NVLink connectivity, memory architecture, packaging, manufacturing and rack-scale integration needed for production deployment. MediaTek is also part of the broader NVIDIA NVLink Fusion ecosystem that lets partners connect custom XPUs to NVIDIA’s AI infrastructure and use the NVIDIA MGX rack-scale architecture to reduce development complexity and accelerate time to market.

“Together, we’re building platforms that bring NVIDIA accelerated computing to new markets and give customers the freedom to create differentiated AI systems at enormous scale,” said Jensen Huang, Founder and CEO of NVIDIA.

“By combining NVIDIA’s leadership in accelerated computing and AI software ecosystem with MediaTek’s expertise in a diverse AI technology portfolio from edge to cloud, and our leadership position in custom silicon, we can accelerate innovation for our customers,” said Rick Tsai, Vice Chairman and CEO of MediaTek.

Tagged with: